FULLmLL                 package:nem                 R Documentation

_F_u_l_l _m_a_r_g_i_n_a_l _l_i_k_e_l_i_h_o_o_d _o_f _a _p_h_e_n_o_t_y_p_i_c _h_i_e_r_a_r_c_h_y

_D_e_s_c_r_i_p_t_i_o_n:

     The function the full marginal likelihood of a phenotypic
     hierarchy. The full marginal likelihood equals the marginal
     likelihood 'mLL'  averaged over the error probabilities $alpha$
     and $beta$.

_U_s_a_g_e:

     FULLmLL(Phi, D1, D0, a0, b0, a1, b1, Pe, Pm=NULL, lambda=0)

_A_r_g_u_m_e_n_t_s:

     Phi: an adjacency matrix with unit main diagonal

      D1: count matrix: phenotypes x genes. How often did we see an
          effect after interventions?

      D0: count matrix: phenotypes x genes. How often did we NOT see an
          effect after intervention?

a0, b0, a1, b1: Hyperparameters

      Pe: prior of effect positions in the hierarchy. A matrix of size
          phenotypes x genes, where each row contains positive numbers
          summing to 1.

      Pm: prior on model graph (n x n matrix) with entries 0 <=
          priorPhi[i,j] <= 1 describing the probability of an edge
          between gene i and gene j.

  lambda: regularization parameter to incorporate prior assumptions.

_D_e_t_a_i_l_s:

     Additionally to the marginal likelihood introduced in Markowetz et
     al (2005), we can average over the error probabilities $alpha$ and
     $beta$ assuming Beta priors. The parameters of the two Beta priors
     are hyperparameters of the full marginal likelihood score. The
     four hyperparameters fall into two categories:  'a1' and 'b0' are
     weights for observing the predicted state, while 'a0' and 'b1' 
     are weights for observing errors. We suggest setting 'a1=b0' and
     'a0=b1'.  The ratio between the two values should correspond to
     our assessment of the noise level. See the example section for an
     application.  The function 'FULLmLL' is usually called from within
     function 'score'.

_V_a_l_u_e:

     mLL: full marginal likelihood of a model

     pos: posterior distribution of effect positions in the hierarchy

  mappos: maximum aposteriori estimate of effect positions

LLperGene: likelihood per E-gene

_A_u_t_h_o_r(_s):

     Florian Markowetz <URL: http://genomics.princeton.edu/~florian>

_R_e_f_e_r_e_n_c_e_s:

     Markowetz F, Probabilistic Models for Gene Silencing Data. PhD
     thesis, Free University Berlin, 2006.

_S_e_e _A_l_s_o:

     'nem', 'score', 'mLL'

_E_x_a_m_p_l_e_s:

        
        data("BoutrosRNAi2002")
        res <- nem(BoutrosRNAiDiscrete[,9:16],type="FULLmLL",hyperpara=c(1,9,9,1))

